• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.2
• /
• pp.103-108
• /
• 2012

Recently, as the prevalence of high-performance mobile devices and the application of the multimedia content are expanded, Super Resolution (SR) technique which reconstructs low resolution images to high resolution images is becoming important. And in the mobile devices, the development of the SR algorithm considering the operation quantity or memory is required because of using the restricted resources. In this paper, we propose a new single frame fast SR technique suitable for mobile devices. In order to prevent color distortion, we change RGB color domain to HSV color domain and process the brightness information V (Value) considering the characteristics of human visual perception. First, the low resolution image is enlarged by the improved fast back projection considering the noise elimination. And at the same time, the reliable edge map is extracted by using the LoG (Laplacian of Gaussian) filtering. Finally, the high definition picture is reconstructed by using the edge information and the improved back projection result. The proposed technique removes effectually the unnatural artefact which is generated during the super resolution restoration, and the edge information which can be lost is amended and emphasized. The experimental results indicate that the proposed algorithm provides better performance than conventional back projection and interpolation methods.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.4
• /
• pp.23-31
• /
• 2012

This paper proposes a motion estimation-based super resolution algorithm to restore input low-resolution images of large movement into a super-resolution image. It is difficult to find the sub-pixel motion estimation in images of large movement compared to typical experimental images. Also, it has disadvantage which have high computational complexity to find reference images and candidate images using general motion estimation method. In order to solve these problems for the traditional two-dimensional motion estimation using the proposed registration threshold that satisfy the conditions based on the reference image is determined. Candidate image with minimum weight among the best candidates for super resolution images, the restoration process to proceed with to find a new image registration algorithm is proposed. According to experimental results, the average PSNR of the proposed algorithm is 31.89dB and this is better than PSNR of traditional super-resolution algorithm and it also shows improvement of computational complexity.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.2
• /
• pp.70.4-70.4
• /
• 2019

Image super-resolution (SR) is a technique that enhances the resolution of a low resolution image. In this study, we use three SR models (RCAN, ProSRGAN and Bicubic) for enhancing solar SDO/HMI magnetograms using deep learning. Each model generates a high resolution HMI image from a low resolution HMI image (4 by 4 binning). The pixel resolution of HMI is about 0.504 arcsec. Deep learning networks try to find the hidden equation between low resolution image and high resolution image from given input and the corresponding output image. In this study, we trained three models with HMI images in 2014 and test them with HMI images in 2015. We find that the RCAN model achieves higher quality results than the other two methods in view of both visual aspects and metrics: 31.40 peak signal-to-noise ratio(PSNR), Correlation Coefficient (0.96), Root mean square error (RMSE) is 0.004. This result is also much better than the conventional bi-cubic interpolation. We apply this model to a full-resolution SDO/HMI image and compare the generated image with the corresponding Hinode NFI magnetogram. As a result, we get a very high correlation (0.92) between the generated SR magnetogram and the Hinode one.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.104-123
• /
• 2022

Style transfer based on neural network provides very high quality results by reflecting the high level structural characteristics of images, and thereby has recently attracted great attention. This paper deals with the problem of resolution limitation due to GPU memory in performing such neural style transfer. We can expect that the gradient operation for style transfer based on partial image, with the aid of the fixed size of receptive field, can produce the same result as the gradient operation using the entire image. Based on this idea, each component of the style transfer loss function is analyzed in this paper to obtain the necessary conditions for partitioning and padding, and to identify, among the information required for gradient calculation, the one that depends on the entire input. By structuring such information for using it as auxiliary constant input for partition-based gradient calculation, this paper develops a recursive algorithm for super high-resolution image style transfer. Since the proposed method performs style transfer by partitioning input image into the size that a GPU can handle, it can perform style transfer without the limit of the input image resolution accompanied by the GPU memory size. With the aid of such super high-resolution support, the proposed method can provide a unique style characteristics of detailed area which can only be appreciated in super high-resolution style transfer.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.9
• /
• pp.3745-3761
• /
• 2020

Low-rank representation methods already achieve many applications in the image reconstruction. However, for high-gradient image patches with rich texture details and strong edge information, it is difficult to find sufficient similar patches. Existing low-rank representation methods usually destroy image critical details and fail to preserve edge structure. In order to promote the performance, a new representation-based image super-resolution reconstruction method is proposed, which combines gradient domain guided image filter with the structure-constrained low-rank representation so as to enhance image details as well as reveal the intrinsic structure of an input image. Firstly, we extract the gradient domain guided filter of each atom in high resolution dictionary in order to acquire high-frequency prior information. Secondly, this prior information is taken as a structure constraint and introduced into the low-rank representation framework to develop a new model so as to maintain the edges of reconstructed image. Thirdly, the approximate optimal solution of the model is solved through alternating direction method of multipliers. After that, experiments are performed and results show that the proposed algorithm has higher performances than conventional state-of-the-art algorithms in both quantitative and qualitative aspects.

• Broadcasting and Media Magazine
• /
• v.25 no.2
• /
• pp.7-16
• /
• 2020

최근 Ultra-High Definition(UHD) 등의 고해상도 디스플레이가 시장에 등장하고 이에 따라 소비자의 요구가 커지면서 기존의 Full-High Definition(FHD)과 같은 저해상도(Low Resolution, LR) 영상을 고해상도(High Resolution, HR) 영상으로 변환할 수 있는 초해상화(Super-Resolution) 알고리즘에 대한 관심이 커지고 있다. 그러나 기존의 전통적인 초해상화 기법들은 고해상도 영상을 복원하는 과정에서 디테일한 부분의 화질 저화 및 열화가 존재하는 것을 확인할 수 있다. 본 논문에서는 이러한 문제를 해결하기 위해 최근 널리 연구되고 있는 딥러닝 기반의 초해상화 알고리즘 연구의 최신 기술 현황 및 동향을 소개하고자 한다. 딥러닝 기반의 초해상화 알고리즘은 기존의 전통적인 기법들에 비해 높은 성능을 보여주고 있으며 현재까지도 다양한 네트워크가 제안되며 활발히 연구되고 있다. 본 논문에서는 딥러닝 기반의 초해상화 알고리즘 중 대표적인 네트워크들을 분석하고 다양한 데이터 셋에 대한 해당 네트워크의 실험 결과를 통해 딥러닝 기반의 초해상화 알고리즘의 우수성을 확인하고자 한다.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.63-70
• /
• 2021

Generating a super-resolution meteological data by using a high-resolution deep neural network can provide precise research and useful real-life services. We propose a new technique of generating improved training data for super-resolution deep neural networks. To generate high-resolution meteorological data with domain specific knowledge, Lambert conformal conic projection and objective analysis were applied based on observation data and ERA5 reanalysis field data of specialized institutions. As a result, temperature and humidity analysis data based on domain specific knowledge showed improved RMSE by up to 42% and 46%, respectively. Next, a super-resolution generative adversarial network (SRGAN) which is one of the aritifial intelligence techniques was used to automate the manual data generation technique using damain specific techniques as described above. Experiments were conducted to generate high-resolution data with 1 km resolution from global model data with 10 km resolution. Finally, the results generated with SRGAN have a higher resoltuion than the global model input data, and showed a similar analysis pattern to the manually generated high-resolution analysis data, but also showed a smooth boundary.

• Journal of Korea Multimedia Society
• /
• v.17 no.12
• /
• pp.1393-1401
• /
• 2014

This paper presents an adaptive image interpolation method using a pixel-based neighbor embedding which is modified from the patch-based neighbor embedding of contemporary super resolution algorithms. Conventional interpolation methods for high resolution detect at least 16-directional edges in order to remove zig-zaging effects and selectively choose the interpolation strategy according to the direction and value of edge. Thus, they require much computation and high complexity. In order to develop a simple interpolation method preserving edge's directional shape, the proposed algorithm adopts the simplest Haar wavelet and suggests a new pixel-based embedding scheme. First, the low-quality image but high resolution, magnified into 1 octave above, is acquired using an adaptive 8-directional interpolation based on the high frequency coefficients of the wavelet transform. Thereafter, the pixel embedding process updates a high resolution pixel of the magnified image with the weighted sum of the best matched pixel value, which is searched at its low resolution image. As the results, the proposed scheme is simple and removes zig-zaging effects without any additional process.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.2
• /
• pp.117-137
• /
• 2010

Iris recognition is a biometric technique which uses unique iris patterns between the pupil and sclera. The advantage of iris recognition lies in high recognition accuracy; however, for good performance, it requires the diameter of the iris to be greater than 200 pixels in an input image. So, a conventional iris system uses a camera with a costly and bulky zoom lens. To overcome this problem, we propose a new method to restore a low resolution iris image into a high resolution image using a single image. This study has three novelties compared to previous works: (i) To obtain a high resolution iris image, we only use a single iris image. This can solve the problems of conventional restoration methods with multiple images, which need considerable processing time for image capturing and registration. (ii) By using bilinear interpolation and a constrained least squares (CLS) filter based on the degradation model, we obtain a high resolution iris image with high recognition performance at fast speed. (iii) We select the optimized parameters of the CLS filter and degradation model according to the zoom factor of the image in terms of recognition accuracy. Experimental results showed that the accuracy of iris recognition was enhanced using the proposed method.

• Transactions of the Society of Information Storage Systems
• /
• v.11 no.2
• /
• pp.22-25
• /
• 2015

In this paper, we demonstrate solid-immersion lens based confocal microscopy using super-continuum generation effect. Using super-continuum generation effect, we could diversify the excitation wavelength of confocal microscopy. Further, high refractive index of solid-immersion lens would increase the resolution of confocal microscopy. As a result, by applying the super-continuum generation effect and solid-immersion lens to confocal microscopy, some problems of confocal fluorescent microscopy, the excitation wavelength and the resolution, could be overcome. To verify it, we made home-built solid-immersion lens based confocal microscopy using super-continuum generation effect, and evaluate the performance of the system.